Medulloblastoma (MB) is the most common malignant brain tumor in children. Although surgery, radiation and high-dose chemotherapy have led to increased survival rates, many MB patients still die from their disease, in part due to the high incidence of tumor recurrence. Thus, novel and more effective therapies for MB are critically important. The most aggressive subtype of MB consists of tumors that exhibit overexpression or amplification of the MYC oncogene. Patients with MYC-driven MB initially to respond to therapy, but then relapse months or years later, suggesting that some tumor cells can persist, and remain dormant for prolonged periods after treatment. However, the molecular basis of this dormancy, and the events that lead to tumor recurrence, remain unclear. We recently developed a unique model of MYC-driven MB that allows us to induce dormancy, to identify and isolate dormant tumor cells (DTCs), and to compare these cells to the normal stem cells from which they arise. Using this model, we propose to: (1) Determine whether DTCs resemble stem cells phenotypically and functionally, (2) Identify the molecular mechanisms of tumor cell dormancy by comparing the gene expression profiles of DTCs, stem cells and malignant tumor cells, and (3) Screen for small molecules that promote differentiation of dormant tumor cells, but are not toxic to normal stem cells. If these studies are successful, they will help identify novel approaches to preventing tumor relapse in MYC-driven MB and other malignancies. !